Bi-spectral retrievals of droplet effective radius and cloud optical depth are widely utilized to estimate aerosol cloud interactions (ACI) in warm clouds in the marine boundary layer. Here, we assess the effect of retrieval errors due to the neglect of 3D radiative transfer during the retrieval process on this analysis of ACI. We use an ensemble of stochastically-modelled cloud fields and 3D radiative transfer simulations to study the retrieval errors at a solar zenith angle of 30. Simulated retrieval biases in droplet number concentration (𝑵𝒅) for all three MODIS channels vary systematically from +35% to -80% as cloud heterogeneity increases. Pixel-level errors can be much larger. Commonly utilized subsampling strategies do not reduce the systematic variation in retrieval error. Negative error correlations between optical depth and droplet effective radius produce spuriously negative slopes between the logarithm of liquid water path and 𝑵𝒅 (-1.0 to -0.3). Pixels at the centre of (8 km)2 patches that are not overcast have a relative bias in 𝑵𝒅 of -50%. The relative frequency of these biased pixels varies linearly with the clear fraction in MODIS data and form the basis for a simple parameterization of 𝑵𝒅 bias that varies with cloud fraction. Using this parameterization, synthetic data experiments indicate that estimates of the first indirect effect in the tropical ocean can be overestimated by up to 30%, and the effect of aerosol on cloud fraction is overestimated by 50%, due to neglecting the correlation between retrieval errors and cloud microphysics.

Our longest, stable record of cloud-top pressure (CTP) and cloud-top height (CTH) are derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Multi-Angle Imaging Spectroradiometer (MISR) on Terra. Because of single cloud-layer assumptions in their standard algorithms, they provide only single CTP/CTH retrievals in multi-layered situations. In the predominant multi-layered regime of thin cirrus over low clouds, MODIS significantly overestimates cirrus CTP and emissivity, while MISR accurately retrieves low-cloud CTH. Utilizing these complementary capabilities, we develop a retrieval algorithm for accurately determining both-layer CTP and cirrus emissivity for such 2-layered clouds, by applying the MISR low-cloud CTH as a boundary condition to a modified MODIS CO2-slicing retrieval. We evaluate our 2-layered retrievals against collocated Cloud-Aerosol Transport System (CATS) lidar observations. Relative to CATS, the mean bias of the upper cloud CTP and emissivity are reduced by ~90% and ~75% respectively in the new technique, compared to standard MODIS products. We develop an error model for the 2-layered retrieval accounting for systematic and random errors. We find up to 88% of all residuals lie within modeled 95% confidence intervals, indicating a near-closure of error budget. This reduction in error leads to a reduction in modeled atmospheric longwave radiative flux biases ranging between 5-40 Wm-2, depending on the position and optical properties of the layers. Given this large radiative impact, we recommend that the pixel-level 2-layered MODIS+MISR fusion algorithm be applied over the entire MISR swath for the 22-year Terra record, leading to a first-of-its-kind 2-layered cloud climatology from Terra’s morning orbit.